基于可拓距K-均值聚类和正弦微分进化算法的风储联合系统优化配置

以最大化风储联合投资商的收益和风电就地消纳率为目标,考虑源网荷协同优化和需求响应构建了风储联合系统的多目标优化配置模型;采用可拓距K-均值聚类算法对分布式风电出力和负荷需求的不确定性进行多场景分析,以实现更为准确而均衡的场景缩减;通过引入多核并行运行环境与正弦函数的思想,提出基于并行多目标正弦微分进化算法对优化配置模型进行高效求解;以IEEE 33节点配电系统为算例进行风储联合系统的优化配置,仿真结果验证了所建模型的有效性和优越性.     

GM模型的几何性质及统计检验

以空间向量几何理论为基础，详细地分析了Ｇａｕｓｓ－Ｍａｒｋｏｖ线性模型的几何性质；在实现了模型图形化描述的基础上，利用各解向量的几何及统计性质构造出了不同情形下的统计检验方法和相应的检验统计量，为全面、深入地研究和灵活运用该模型提出了一种简洁直观的数学方法     

热轧钢卷强制冷却工艺的研究

试验研究了热轧钢卷的强制冷却工艺过程 ,探讨了钢卷强制冷却工艺参数对带钢表面质量、力学性能及板形等的影响 ,认为卷取后采用间断式强冷工艺可获得较好的表面质量、较均匀的力学性能 ,并能明显缩短钢卷库存时间     

Development of new device for measuring thermal stresses

In recent years, numerous analytical and experimental researches have been performed on the prediction of thermal stresses in mass concrete structures. However, due to the difficulty of the problem, limitations still exist for both analytical and experimental methods of measuring thermal stresses in mass concrete. In this research, a new experimental device measuring thermal stresses directly in a laboratory setting is developed. The equipment is located in a temperature chamber that follows the temperature history, which has been previously obtained from temperature distribution analyses. Thermal forces are measured continuously by two load cells in the device. The results show that the thermal stresses estimated by the newly developed device agree well with general stress variations in actual structures.     

预测控制约束边界效应与解决方法研究

约束预测控制（Constrained model predictive control,CMPC）中,因约束的存在,优化过程中最优控制作用可能会在可行域的边界取值,也就是说会 有一个或多个变量饱和,即约束边界效应. 而过程控制中操纵变量饱和是我们不希望出现的. 对此,首先基于稳态模型,对期望值位于可行域内时最优解必在期望值处达到给出证明;同时证明了期望值在可行域外时最优解可转化为期望值到可行域的投影. 其次,针对变量在动态及稳态过程中饱和的情况提出了改善控制性能的措施——调整目标函数;终端约束的加入,为预测控制系统稳定性提供了保障. 通过对包含约束的连续搅拌釜式反应器（Continuous stirred tank reactor,CSTR）系统进行仿真实验,验证了所提方法的正确性,并说明了对目标函数进行适当调整,可有效改善系统的控制性能.     

路径长度受限的随机需求VRP的模型和算法

路径长度受限的随机需求VRP在生产、生活中有着广泛的应用。给出了路径长度受限的随机需求VRP问题的线性整数规划模型,在不允许部分服务和仅能服务失败一次的策略下设计了一个启发式算法。该算法以离散优化中广泛应用的扫描算法为基础生成服务的一个初始可行方案,然后利用模拟退火算法改进得到近似最优解。对需求为二项分布的50个结点、1个服务中心的问题进行了数值试验,数值结果表明该算法对求解路径长度受限的随机需求VRP是有效的。     

Formulation of unified Jacobian for serial-parallel manipulators

This paper presents a general approach for Jacobian analysis of serial-parallel manipulators (S-PMs) formed by two lower mobility parallel manipulators (PMs) connected in serials. Based on the kinematic relation, coupling and constraint properties of each PM, the unified forward and inverse Jacobian matrices for S-PMs are derived in explicit form. It is shown that the Jacobian matrices of S-PMs have unified forms, which include the complete information of each PM. A (3-RPS)+(3-SPS/UP) S-PM is used as an example to demonstrate the proposed approach. The established model is applicable for S-PMs with various architectures.     

Synthesis and chain extension of nitroxide‐terminated styrene–maleimide copolymers

Thermal radical copolymerization of styrene (S) and maleimide (MI) at 125°C in diglyme in the presence of 2,2,6,6‐tetramethylpiperidin‐1‐yloxyl radical (TEMPO) was studied. Mole fractions of maleimide in the feed, F_MI, varied in the range 0.1–0.9. A quasiliving reaction process proceeded yielding copolymers with a low polydispersity (M_w/M_n = 1.17–1.41). The found azeotropic composition, (F_MI)_A = 0.46, did not differ substantially from that (0.5) in the conventional radical S‐MI copolymerization. At a higher conversion or MI content in the feed, deactivation of the copolymer chains occurred. The obtained TEMPO‐terminated S‐MI copolymers readily initiated polymerization of styrene; chain extension of the macroinitiators took place, giving poly(S‐co‐MI)‐block‐poly(S) diblock copolymers. The synthesized copolymers containing S and MI units were characterized by elemental analysis, NMR spectroscopy, size‐exclusion chromatography, and differential scanning calorimetry. © 2004 Wiley Periodicals, Inc. J Appl Polym Sci 92: 1863–1868, 2004     

Credit rating with a monotonicity-constrained support vector machine model

Deciding whether borrowers can fulfill their obligations is a major issue for financial institutions, and while various credit rating models have been developed to help achieve this, they cannot reflect the domain knowledge of human experts. This paper proposes a new rating model based on a support vector machine with monotonicity constraints derived from the prior knowledge of financial experts. Experiments conducted on real-world data sets show that the proposed method, not only data driven but also domain knowledge oriented, can help correct the loss of monotonicity in data occurring during the collecting process, and performs better than the conventional counterpart.     

On the performance of economic model predictive control with self-tuning terminal cost

In this paper, we analyze the closed-loop performance of a recently introduced economic model predictive control (MPC) scheme with self-tuning terminal cost. To this end, we propose to use a generalized terminal region constraint instead of a generalized terminal equality constraint within the repeatedly solved optimization problem, which allows us to obtain improved closed-loop asymptotic average performance bounds. In particular, these bounds can be obtained a priori. We discuss how the necessary parameters for the generalized terminal region setting can be calculated, and we illustrate our findings with two numerical examples.